Promoting Clean Energy in the American Power Sector

Transcription

1 Promoting Clean Energy in the American Power Sector The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters Citation Aldy, Joseph Promoting Clean Energy in the American Power Sector. Hamilton Project Discussion Paper , Brookings Institution Published Version Citable link Terms of Use This article was downloaded from Harvard University s DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at nrs.harvard.edu/urn-3:hul.instrepos:dash.current.terms-ofuse#laa

2 DISCUSSION PAPER MAY 2011 Promoting Clean Energy in the American Power Sector Joseph E. Aldy

3 MISSION STATEMENT The Hamilton Project seeks to advance America s promise of opportunity, prosperity, and growth. The Project s economic strategy reflects a judgment that long-term prosperity is best achieved by fostering economic growth and broad participation in that growth, by enhancing individual economic security, and by embracing a role for effective government in making needed public investments. We believe that today s increasingly competitive global economy requires public policy ideas commensurate with the challenges of the 21st Century. Our strategy calls for combining increased public investments in key growth-enhancing areas, a secure social safety net, and fiscal discipline. In that framework, the Project puts forward innovative proposals from leading economic thinkers based on credible evidence and experience, not ideology or doctrine to introduce new and effective policy options into the national debate. The Project is named after Alexander Hamilton, the nation s first treasury secretary, who laid the foundation for the modern American economy. Consistent with the guiding principles of the Project, Hamilton stood for sound fiscal policy, believed that broad-based opportunity for advancement would drive American economic growth, and recognized that prudent aids and encouragements on the part of government are necessary to enhance and guide market forces.

4 Promoting Clean Energy in the American Power Sector Joseph E. Aldy John F. Kennedy School of Government, Harvard University Resources for the Future National Bureau of Economic Research MAY 2011 NOTE: This discussion paper is a proposal from the author. As emphasized in The Hamilton Project s original strategy paper, the Project was designed in part to provide a forum for leading thinkers across the nation to put forward innovative and potentially important economic policy ideas that share the Project s broad goals of promoting economic growth, broad-based participation in growth, and economic security. The authors are invited to express their own ideas in discussion papers, whether or not the Project s staff or advisory council agrees with the specific proposals. This discussion paper is offered in that spirit. The Hamilton Project Brookings 1

5 Abstract Despite bipartisan interest in advancing American energy policy, comprehensive energy and climate legislation fell short in the Senate last year after passing in the House of Representatives in The difficulty of coming to broad agreement highlights the need for a more targeted and incremental approach. One promising intermediate step would be a technology-neutral national clean energy standard that applies to the U.S. power sector. This paper proposes a standard that would lower carbon dioxide emissions by as much as 60 percent relative to 2005 levels over twenty years, streamline the fragmented regulatory system that is currently in place, generate fiscal benefits, and help fund energy innovation. Through a simple design and transparent implementation, the National Clean Energy Standard would provide certainty about the economic returns to clean energy that would facilitate investment in new energy projects and lower the emission intensity of the power sector. It would also serve as an ambitious bridge to economy-wide energy and climate policy. 2 Promoting Clean Energy in the American Power Sector

6 Table of Contents ABSTRACT 2 INTRODUCTION 5 CHAPTER 1: RELIANCE ON DIRTY POWER 7 CHAPTER 2: DETAILED PROPOSAL 11 CHAPTER 3: RESPONSES TO SOME LIKELY QUESTIONS 20 CONCLUSION 27 AUTHOR AND ACKNOWLEDGMENTS 28 ENDNOTES 29 REFERENCES 30 The Hamilton Project Brookings 3

7 Introduction The general public and their representatives in Washington from both sides of the political aisle agree that the United States should advance a more sensible energy policy. Under the status quo, the U.S. power sector is characterized by dirty power and a complex and potentially costly regulatory environment. The U.S. power sector emits more carbon dioxide (CO 2 ) than any other country in the world except China. Continuing climate science scholarship highlights the need to take action to mitigate the risk posed by greenhouse gas emissions. Deploying clean energy in the power sector can contribute to the global effort to combat climate change while providing demand for domestic natural gas production and manufacturing and construction associated with new zero-emission power facilities, including renewables and nuclear. About thirty states have some form of renewable or clean energy mandate in their power sectors. This illustrates significant interest across the country in both blue and red states to promote clean energy. In the past year, interest in taking a more expansive approach to promoting clean energy has been evident from both Democrats and Republicans: Senators Lugar, Graham, and Murkowski supported a bill with a diverse energy standard (S. 3464) and President Obama called for a clean energy standard. 1 A national policy could lower bureaucratic burdens of the state regulatory patchwork for those operating in multiple states and deliver societal goals in a more cost-effective manner. Layered on top of this is the prospect for federal greenhouse gas regulations under the Clean Air Act. This Clean Air Act authority is unlikely to result in an approach that effectively promotes clean energy and lowers greenhouse gas emissions at a cost that is justified by its benefits. Well-designed policy requiring new legislation is necessary. A preferred energy policy should deliver an array of benefits, including reducing the fiscal burden of the energy system, promoting investment that has been inhibited by policy uncertainty, streamlining the regulatory landscape, and improving the global environment. A first-best policy to deliver these objectives would be an economy-wide carbon tax that could finance energy research and development (R&D), tax cuts, and deficit reduction, and replace the existing and potentially complex state and federal regulatory systems. Given the current political challenge in advancing such a first-best policy, I propose a National Clean Energy Standard (NCES) for the U.S. power sector. The principles of simplicity, cost-effectiveness, and price certainty guided the design for the proposed standard that should make meaningful progress on these policy objectives. The key details of the NCES include A technology-neutral performance metric. My definition of clean energy is based on the CO 2 emission intensity of power generation. The lower the intensity, the cleaner the power. This avoids the challenge of picking technology winners that could risk inhibiting future innovation, and makes the assessment of progress transparent and simple. Performance goals. The share of power generated by clean energy technologies would increase over time, reflecting a carbon emission performance standard for the power sector that becomes more stringent through The initial performance goal in 2015 is 0.4 tons of CO 2 per megawatt hour of generation (tco 2 /MWh), which would require meaningful improvement from the 2010 emission The principles of simplicity, cost-effectiveness, and price certainty guided the design for the proposed standard that should make meaningful progress on key energy policy objectives. The Hamilton Project Brookings 5

8 intensity of generation of This 2015 goal of 0.4 tco 2 / MWh roughly corresponds to a 60 percent clean energy portfolio standard. The performance goal would tighten by 0.01 tco 2 /MWh each year until reaching 0.2 in 2035, approximately equal to an 80 percent clean energy portfolio. Clean energy credits. A power producer that beats the performance goal can create tradable clean energy credits that it may sell to utilities that fail to meet the goal. This would promote cost-effective deployment of clean energy in the power sector. Each credit is denominated in terms of a ton of CO 2. Compliance. A utility may demonstrate compliance with the policy through a combination of approaches. First, it may reduce the intensity of its generation below the performance goal. Second, it may purchase clean energy credits from other utilities such that its performance and purchased credits meet the goal. Third, it may purchase federal clean energy credits that are available at a preset price, such that these in combination with privately generated credits and the utility s own performance meet the goal. In 2015, the price for these credits will be $15, and it will ramp up 7 percent annually, reaching $30 per credit by The federal clean energy credit prices are expected to bind on the clean energy credit market, which would provide price certainty to promote and improve investment decisions and provide a guarantee against unexpectedly high prices. Over the first ten years of the program, the price would average $21/MWh per credit, which would deliver the same value to wind and many other renewable producers as they enjoy today through the production tax credit, without the fiscal outlay. Clean Energy Fund. Revenues raised through the federal clean energy credit would finance a ramp-up in energy R&D. I propose dedicating $2 billion of revenues in 2015 to the fund, and ramping this up over time to $5 billion in The fund should support merit-based, competitive research, such as through the Advanced Research Projects Agency Energy (ARPA-E), and could also support first-ofa-kind demonstration projects as outlined in Deutch (2011) as well as other energy R&D programs. Any excess revenues beyond this would be dedicated to deficit reduction or reducing current tax rates, such as the payroll tax rate. Environmental Protection Agency (EPA) Greenhouse Gas Regulatory Authority and State Renewable Mandates. The NCES would eliminate the need for EPA regulations of greenhouse gas emissions in the power sector, and the need for significantly duplicative state renewable mandates. Legislation creating a NCES should strike EPA greenhouse gas regulatory authority in the power sector and preempt state renewable and alternative energy portfolio standards. The federal clean energy credit prices... would provide price certainty to promote and improve investment decisions and provide a guarantee against unexpectedly high prices. 6 Promoting Clean Energy in the American Power Sector

9 Chapter 1: Reliance on Dirty Power A by-product of America s reliance on power produced from relatively dirty fuel is relatively high carbon emissions. On average, a megawatt hour of electricity generated in the United States today results in emissions of about 0.56 metric tons of CO 2. Whereas the evolving mix of power generation technologies has resulted in a declining emission intensity of electricity in the post-war period (Figure 1), it remains well above the intensities of Japan and the European Union. The modest reduction in the emission intensity of the power sector over time reflects both improvements in combustion efficiency and changes in the fuel mix. Emission intensity of power generation varies significantly across technologies and fuels. Natural gas produces less than half as much carbon pollution per unit of electricity generated as does coal, and nuclear and renewable sources have zero CO 2 emissions (Table 1). 2 Within a given type of fossil fuel power plant, emission intensity can vary with the facility s combustion efficiency and fuel characteristics. Successful commercialization of carbon capture and storage technologies could dramatically alter the emission profiles for fossil fuel based generation technologies. FIGURE 1 U.S. Electricity Generation Carbon Dioxide Emission Intensity, Source: Constructed from data presented in EIA (2010a, 2011b). Note: The 2010 emission intensity is estimated based on 2010 fuel shares and 2009 emission intensities by type of generation. The Hamilton Project Brookings 7

10 TABLE 1 Average Carbon Dioxide Emission Intensity by Generation Type, 2009, tco 2 /MWh Coal Oil Natural Gas Nuclear Hydroelectric Other renewables Source: Constructed from data presented in EIA (2010a). Coal has served as the leading source of power in the United States since World War II, ranging between about 44 and 57 percent of U.S. power generation (Figure 2), although its share of power generation in 2009 and 2010 was lower than at any point since the late 1970s. The other primary baseload power source, nuclear, experienced an increase in its share of power generation in the 1970s as new power plants came online. Nuclear has represented about one-fifth of U.S. power over the past two decades. Refined petroleum products, primarily residual fuel, distillate fuel, and petroleum coke once comprised a meaningful share of U.S. power generation. At the time of the first oil shock in 1973, oil-fired power plants generated 17 percent of U.S. power and represented about 10 percent of U.S. oil consumption. By 2010, oil produced less than 1 percent of U.S. generation; the total number of barrels of oil consumed by the power sector has fallen 90 percent since Given oil s very small share of generation today, the environmental and energy security benefits of further efforts to back oil out of the power sector are quite modest. 3 The share of power from natural gas plants has varied significantly since In 1970, nearly one-quarter of U.S. electricity came from gas-fired plants. By 1990, the gas share of power generation had fallen to about 10 percent, but has since rebounded to exceed 20 percent over the past four years. The gas share of generation in 2010 was the highest it has been since Renewable power, primarily through hydroelectric dams, delivered more than 20 percent of U.S. power in the 1950s, but hydroelectric s share has declined because of limited investment in dams since then. Other renewable generation sources, including wind, solar, geothermal, and biomass, made up no more than about 1 to 1.5 percent of power generation in the 1990s. The share of other renewables doubled from 2006 to 2010, and is now about 3.5 percent. FIGURE 2 Share of U.S. Electricity Generation by Type, Source: Constructed from data presented in EIA (2010a, 2011b). 8 Promoting Clean Energy in the American Power Sector

11 The shift toward natural gas and renewable power had a material impact on U.S. power sector carbon emissions in 2009, in addition to the effect of declining demand resulting from lower economic activity. If the share of power generated from coal, natural gas, and oil did not change from 2008 to 2009, then electricity sector CO 2 emissions would have been about 5 percent higher (about 100 million metric tons higher) than they were in The increase in power generation from natural gas and other renewables reflects the substantial increase in investment in generating capacity over the past two decades. During the period , total U.S. generating capacity increased by about 276 gigawatts, representing growth of more than 37 percent (Table 2). Coal-generating capacity changed by only 6 gigawatts, or about 2 percent of the growth in capacity. In contrast, natural gas capacity increased by more than 250 gigawatts and made up about 92 percent of the growth in total capacity. Other renewable capacity has increased significantly, especially in recent years. By 2010, wind capacity exceeded 40 gigawatts (American Wind Energy Association [AWEA] 2011). Nameplate-generating capacity does not correspond oneto-one to power generation. Some power plant technologies are intended for baseload, while some are intended to provide power during peak periods of demand. In addition, intermittent sources, such as solar and wind, produce less power per megawatt of installed capacity than baseload coal and nuclear. The Energy Information Administration (EIA 2011a) published estimates of capacity factors for new generating capacity that could come online within five years (Table 3). These estimates show that a megawatt of capacity of combined-cycle natural gas, in contrast to combustion turbine natural gas plants built primarily to meet peak demand, can compete with a megawatt of coal or nuclear in serving a given base demand. About two-thirds of the new natural gas generating capacity to come online since 1990 is combined cycle, but high natural gas prices over limited its use (Kaplan 2010). For example, only 13 percent of the combined-cycle natural gas plants in operation in 2007 had capacity factors that exceeded 70 percent. The average capacity factor for coal in 2007 was 75 percent, significantly higher than the average capacity factor of 42 percent for combined-cycle natural gas in that year (Kaplan 2010). TABLE 2 U.S. Electricity Summer Generating Capacity, Gigawatts, 1990 and 2008 Type Total 734 1,010 Coal Natural gas Other renewable 2 25 Source: Constructed from data presented in EIA (2010a). Note: Other fuels experienced modest changes from their 1990 levels, with the exception of oil-fired capacity, which declined about 20 gigawatts. TABLE 3 Estimated Capacity Factor of New Generating Capacity for 2016 Conventional Combined- Advanced Wind Solar Geothermal coal cycle natural gas nuclear 85% 87% 90% 34% 18 25% 92% Source: EIA (2011a). Note: Solar range represents solar thermal (18 percent) and solar PV (25 percent). The Hamilton Project Brookings 9

12 The recent increase in power generation from natural gas plants since 2008 represents increasing utilization of existing facilities. Given the forecast for slow growth in electricity demand (less than 1 percent per year through 2035) and the existing surplus of natural gas fired generation capacity, the EIA (2011a) forecasts modest investment in new power plant capacity over the next two decades. Through 2030, cumulative electricity capacity investments are forecast to total about 125 gigawatts to meet slightly higher demand and offset retirements of about 35 gigawatts of capacity. The mix in power generation is expected to evolve slowly, with a modest increase in renewable and natural gas over time, and a slightly lower emission intensity of generation over the next several decades (Figure 3). FIGURE 3 Forecast Power Generation Shares and Carbon Dioxide Emission Intensity, Source: Constructed from EIA (2011a). 10 Promoting Clean Energy in the American Power Sector

13 Chapter 2: Detailed Proposal A well-designed policy to promote the deployment of clean energy in the power sector can help address fiscal challenges, encourage investment, streamline the regulatory landscape, and provide environmental benefits. OBJECTIVES Delivering fiscal benefits. A clean energy standard creates demand-side incentives for clean energy that may substitute for the current supply-side subsidies through tax credits and grants. This would enable legislators to allocate future resources to energy programs that complement the clean energy standard (such as R&D on basic energy sciences), to other socially desired policies, or deficit reduction in lieu of clean energy deployment subsidies. A clean energy standard can also generate revenues that could be used to finance energy R&D as well as to reduce the deficit or finance reductions in existing tax rates. incentives that investors can use as the basis for structuring financing for new power generation projects. Streamlining regulatory landscape. In addition to resolving investment uncertainty regarding the form and implementation of federal policy, a clean energy standard can establish a single national standard to replace the various state renewable and clean energy mandates. Given the significant variation in the design and implementation of state mandates and the prospect of EPA greenhouse gas regulation, a single, streamlined federal clean energy policy can reduce regulatory complexity and promote clean energy deployment costeffectively. Improving the global environment. Finally, a clean energy standard can drive investment and energy deployment decisions that will lower U.S. power sector greenhouse gas emissions and represent a meaningful step in combating the risks of climate change. Promoting investment. A clean energy standard would reduce the regulatory uncertainty that could be chilling investment in the power sector. This uncertainty takes two forms. First, power companies face uncertainty over the general regulatory framework. Second, some regulatory approaches are characterized by more uncertainty than others from the perspective of those making investment decisions. For example, moving forward with power sector regulations under the Clean Air Act could address the first form of uncertainty, but may not address the latter form because of the prospect of extended litigation (presuming the regulations survive congressional review) and potential variation and uncertainty in state implementation programs. Moreover, a well-designed clean energy policy should deliver transparent, stable...a well-designed clean energy policy should deliver transparent, stable incentives that PRINCIPLES investors can use as the basis for structuring financing for new power generation projects. In order to realize these policy objectives, three key principles should guide the design of a clean energy standard: simplicity, cost-effectiveness, and price certainty. Simplicity. Clean energy can be defined along a variety of dimensions, such as its effects on air pollution, water pollution, toxics, resource throughput, solid and hazardous waste, greenhouse gas emissions; alternatively, it can be defined by some predetermined categorization of technologies. A complex set of measures characterizing how clean a given energy source may be could create regulatory confusion and inhibit investment. A predetermined class of qualifying technologies could inhibit innovation. A simple, transparent policy would minimize the risk of a bumpy and costly start-up to the program. Cost-effectiveness. The mandate should be implemented in a way that promotes deployment of clean energy technologies at the lowest possible cost. The Hamilton Project Brookings 11

14 Price certainty. Investment in energy technologies is enhanced when project developers have a higher degree of certainty regarding the costs and benefits of their investment. Price certainty can also mitigate the risk of unexpectedly and excessively high costs to electricity producers and consumers. POLICY DESIGN Given these objectives and principles, an NCES should take the following form. Technology-neutral metric. The proposed NCES would evaluate performance based on the emission intensity of power measured by the tons of CO 2 per megawatt hour of generation. The lower the emission intensity of generation, the cleaner the technology. An emission intensity-based standard has several virtues. First, it is a simple metric that is based on tons of CO 2 and megawatt hours of generation, two measures that are already monitored at and reported by U.S. power plants. Second, it focuses on the environmental impact of power generation that has not been subject to any regulatory policy, in contrast to conventional air pollution, water pollution, and so on. Third, there is a rough correlation between greenhouse gas emissions and other dirty attributes of power generation by fuel source. Thus, greenhouse gas emissions may serve as a proxy for these other attributes. Fourth, a technology-neutral approach provides incentives to all power plant operators to seek out and exploit low-cost ways to clean up their generating capacity. It does not rule out any class of technologies, as a renewable energy standard would, and thus it avoids the problem of the government choosing winners. The private sector, through its innovation and commercialization of technologies and processes, will identify the winners. A technology-neutral approach also avoids the challenge of evaluating new technologies that may not cleanly fit within a previously established classification scheme. A new technology simply demonstrates its cleanliness in terms of the emission intensity of its generation. Performance goals. To ensure that the share of power generated by clean energy technologies increases and the power sector produces lower CO 2 emissions over time, performance goals should be set to drive investment and deployment of these technologies. Given the 2010 U.S. emission intensity of generation of about 0.56 tons of CO 2 per megawatt hour and the near-term forecast for modest improvement in this measure over the next five years to reach about 0.53 tco 2 / MWh, this proposal calls for a performance standard of 0.40 tco 2 /MWh in The performance would become more ambitious over time, declining by 0.01 tco 2 /MWh per year through 2035, as shown in Table 4. TABLE 4 National Clean Energy Standard Goals through 2035 Year National Clean Energy Standard (tco 2 /MWh) Such standards would drive the U.S. power sector to surpass the current extent of clean energy generation in the European Union and Japan. It would require the U.S. power sector to ramp up the generation of cleaner power at approximately the same annual rate over the next five years as China realized over Point of regulation. The point of regulation for the NCES would be at the power plant. This would take advantage of the existing monitoring system on fossil power units, replicate the current point of regulation for other power plant policies (such as under the Acid Rain Program), and obviate the need to map electricity distributed to consumers back to original generating units that would be required by establishing the point of regulation downstream from the generating facility. 4 This approach also facilitates expansion of the program to cover on-site power generation at manufacturing facilities to reduce the incentive for leakage in other words, the relocation of power generation away from regulated utilities and in an unregulated space. To avoid imposing costs on very small power generators in the manufacturing sector, the EPA and Department of Energy (DOE) should jointly establish a minimum size that triggers coverage by the performance standard through rule making. Legislation creating a clean energy standard should provide the regulatory authorities with the flexibility to design rules tailored to combined heat and power facilities to ensure that the clean energy standard does not distort decisions on power and heat production in a way that causes adverse environmental impacts. Tradable clean energy credits. To promote cost-effective attainment of these goals and to create robust incentives for innovation and deployment, power plants that beat the standard create tradable clean energy credits that can be sold to power plants whose emissions per megawatt hour exceed the standard. A clean energy credit is denominated in increments 12 Promoting Clean Energy in the American Power Sector

15 of one metric ton of CO 2. Clean energy credits may be banked for compliance purposes in a future year. This approach lowers the cost of compliance by power plants with high-technology costs by allowing them effectively to finance low-emission generation by other power plants through their purchase of performance credits. It thus provides an economic incentive for investment in zero and low-emission power plants since they can produce a revenue stream through the sale of clean energy credits. Compliance. Power plants would demonstrate compliance with the standard on an annual basis through a three-month true-up period after each compliance year. A power plant could demonstrate compliance with the NCES through a combination of the following approaches. First, the power plant has lesser or equal emissions per megawatt hour than the standard set to drive clean energy deployment. Second, the power plant may purchase clean energy credits from other power plants such that the combination of clean energy credits and the power plant s own performance satisfies the standard. Third, the power plant may also purchase additional clean energy credits from the federal government at a preset price that, in combination with its own generation profile and purchased clean energy credits, would satisfy the NCES. This is similar to the alternative compliance payments in a number of state renewable portfolio standards (RPSs). The federal clean energy credits would initially be set at $15 (in 2010 dollars). The price for these credits would increase over time. They would be indexed to inflation (measured by the GDP deflator) and increase 7 percent annually above the inflation adjustment over the first ten years until reaching about $30 per credit in This ten-year ramp-up in the program would result in a clean energy credit price in 2025 that is on a par with the estimated damages from the carbon pollution associated with power generation (see Interagency Working Group on Social Cost of Carbon, United States Government 2010, Appendix A1). If the clearing price for tradable clean energy credits equals the federal clean energy credit price, zero-carbon power would receive an average price of $21/MWh for generation over the first ten years of the program. This is equal to the current value of the production tax credit for wind, geothermal, and many other renewable [Tradable credits provide] an economic incentive for investment in zero and low-emission power plants since they can produce a revenue stream through the sale of clean energy credits. sources, and exceeds the value of the production tax credit available to the first 6 gigawatts of new nuclear power under the 2005 Energy Policy Act. Beyond 2025, the price of federal clean energy credits would continue to increase at a rate of 2.4 percent per year (in addition to inflation) to match the estimated increase in the incremental damages from carbon pollution. Clean Energy Fund. An initial tranche of annual revenues generated by the sale of federal clean energy credits would be dedicated to supporting federal energy research. In 2015, the first $2 billion of annual revenues would be directed to support energy R&D. This initial tranche would increase at a rate of 10 percent per year to provide additional funding for energy R&D, and thus would ramp up to about $5 billion in Any excess revenues beyond this tranche may be used for deficit reduction or reducing current tax rates, such as the payroll tax. These funds could increase activities at ARPA-E (the new advanced research projects program at the DOE), finance first-of-a-kind demonstration projects as suggested by Deutch (2011), and address other pressing energy research needs. 5 Implementing agencies. The EPA and the DOE would jointly administer the NCES. This would reflect EPA s experience in implementing cost-effective, market-based approaches to reducing power-generation pollution as well as EPA s existing emission monitoring infrastructure. It also would reflect DOE s experience in monitoring power sector generation and capacity investment and in implementing clean energy R&D programs. ECONOMIC IMPACTS Cost-effectiveness. The envisioned NCES includes several cost-effective design elements intended to minimize the cost of driving clean power generation and of providing environmental benefits. First, the opportunity to buy and sell clean energy credits should enable utilities, entrepreneurs, innovators, and others to seek out and exploit the lowestcost ways of providing clean energy. Second, power plants maintain the option to bank, or save, clean energy credits generated from current clean energy power generation for use in a future compliance period. This will promote dynamic cost-effectiveness. Third, and perhaps most important, the federal clean energy credits available at a preset price provide a guarantee that prices for credits, and hence generation costs net of the NCES and electricity prices, will not exceed specified levels. The Hamilton Project Brookings 13

16 The federal clean energy credit price will very likely determine the clearing price for tradable clean energy credits, given the ambition of the performance goal and the levels of the federal clean energy credit prices, at least in the early years. In doing so, all power plants will face the same incentive to deploy clean energy. Given the increasing stringency in the performance goals over time and the opportunity to bank unused clean energy credits for future use (even if unexpectedly low-cost technologies and fuels come into the power market that might result in compliance at a credit price below the federal credit price) there will be the future demand that can serve to bring the clean energy credit price back to the federal credit price level. Investment certainty. Uncertainty about the magnitude of the incentive for clean energy deployment risks undermining investment. For example, in the context of state renewable energy standards, it is not uncommon for renewable project developers to seek out power purchasing agreements that lock in electricity prices for delivered power over the long term (e.g., twenty years). Volatile prices for state renewable energy Uncertainty about the magnitude of the incentive for clean energy deployment risks undermining investment. credits complicate project financing. In order to secure debt and equity financing, the project developers often have to eliminate this price uncertainty by negotiating a set price for power in the power purchasing agreements with distribution utilities. Designing a clean energy standard that will very likely yield a stream of clean energy credit prices set by the federal credit price over time will provide the kind of certainty that will promote project financing for new power projects. It also will deliver more certainty for those making upstream investment decisions, such as in shale gas development and wind turbine manufacturing capacity, so that decision-makers can understand the potential market demand and returns to such investments. It is also important to recognize that this certainty in the price can have very important implications for the startup of a new policy. Under this proposal, a power plant (or a generation utility operating a number of power plants) considers the time profile of the federal clean energy credit prices and then assesses its opportunities for reducing the emission intensity of generation at a cost below that of the credit. In a policy regime in which credits are tradable and the price of the credits will reflect the eventual clearing price in the credit market, power plants and utilities need to resolve the uncertainty about their own opportunities to deploy clean energy and form expectations about other power plants opportunities to do so, and thus plan against an expected credit price. Some power plants may expect high credit prices and make one set of investments, and other power plants may expect low credit prices and make another set of investments. Given the irreversible nature of some of these investments, these decisions, which ex post may appear as mistakes once the market credit price has been established, may increase the total cost of deploying clean energy relative to this proposal in which power plants do not need to plan against the uncertainty in credit prices. Electricity prices and policy costs. To illustrate the potential impact of this NCES on electricity prices, let us assume that the federal clean energy credit price is binding. Given the variation in electricity regulatory regimes across the nation, I will focus on a simple version of marginal cost pricing to represent competitive markets and a simple version of costof-service (average cost) pricing to represent regulated markets. To provide regional price impacts, I focus on twenty electricity markets as identified by the EIA and for which 2008 and forecast 2015 electricity rates have been published in the Annual Energy Outlook (EIA 2011a). 6 Given the focus on the near term (through 2015), this illustration assumes no changes in emission intensity relative to their 2008 levels and no changes in generating capacity, which yields a potentially upward bias in the electricity rate impacts. In 2015, the federal clean energy credit price will be $15. Power plants would make investments in reducing the emission intensity of their generation until it reached the value set by the federal clean energy credit price. By effectively exempting the first 0.4 tco 2 /MWh, this NCES provides an implicit rate-based output subsidy. Thus, power plants and utilities would adjust generation to reflect this implicit generation subsidy. Changes in electricity rates should reflect both the $15 credit price and the implicit generation subsidy in the NCES. Fischer (2010) and Fischer and Newell (2008) illustrate how, in some cases, the implicit generation subsidy in renewable energy standards and performance-based clean energy standards akin to this proposal could dominate the credit price and effectively result in a net decline in electricity rates in competitive markets. Average costs, and hence rates 14 Promoting Clean Energy in the American Power Sector

17 in cost-of-service regulated markets, would not experience a net decline. To be conservative, I have assumed that electricity prices do not decline under an NCES relative to the forecast business-as-usual levels. In most competitive markets, natural gas is the marginal source of power generation. Given the underutilization of existing combined-cycle natural gas capacity, the marginal source of power in these markets will often have an emission rate on the order of about 0.44 tco 2 /MWh. In some cases, less-efficient combustion turbines may provide power to meet peak demand, and have a higher emission rate, perhaps on the order of about 0.67 tco 2 /MWh. This suggests that the upper-bound price impact with combined-cycle generation technology would be on the order of about 0.06 /kwh, and as much as about 0.4 /kwh with less-efficient combustion turbine technology in 2015, based on this equation: upper-bound retail price impact per megawatt hour = [marginal intensity in 2008 standard] * $15, where the marginal intensity refers to the emission intensity of the marginal generation plant. To be conservative, the estimates presented below reflect an assumption that marginal power is always generated from combustion turbine plants. In regulated markets, a coal-fired power plant with an emission rate of 1.0 tco 2 /MWh does not need clean energy credits to cover the first 40 percent of its emissions in An upper bound on the 2015 average cost impact, and hence the electricity price increase, can then be represented by upper-bound retail price impact per megawatt hour = [average intensity in 2008 standard] * $15, where the average intensity refers to the mean emission intensity of power generation in a given regulated market. The competitive and cost-of-service regulation price impact estimates represent upper bounds because a generator may find low-cost ways to improve the intensity of generation at a unit for less than $15/tCO 2 between 2008 and This could take the form of modifying the fuel mix (changing the type of coal, cofiring with biomass, and so on), investing in more-efficient boiler technology, improving maintenance of existing boiler equipment, and so on. The effect of improving existing units could be more pronounced in cost-of-service markets, since an improvement in every generator impacts the cost impact through the rate-making process, while only improving the marginal generators would impact the electricity rates in competitive markets. Employing this upper bound, illustrative approach to price impacts shows that the average retail price impact in the United States of complying with the proposed NCES is less than 3 percent of the average retail price of electricity forecast for In competitive electricity markets, if combustion turbine gas plants serve as the marginal source of power (a conservative assumption), then compliance would result in an upper-bound estimate of electricity rates about 3 5 percent higher than they are forecast to be in 2015 but still about 8 29 percent lower than they were in In cost-ofservice markets, electricity rates would range between 0 9 percent higher than their 2015 forecast (Figure 4). Fourteen of the twenty electricity markets representing about two out of every three states would have lower electricity prices under this proposed NCES in 2015 than they did in Fourteen of the twenty electricity markets representing about two out of every three states would have lower electricity prices under this proposed NCES in 2015 than they did in The Hamilton Project Brookings 15

18 FIGURE 4 Upper-bound Electricity Rate Impacts (2010 cents per kwh) Region States Code Rate Rate Rate with CES ERCT Texas FRCC Florida MROE Michigan, Wisconsin MROW Iowa, Minnesota, Montana, Nebraska, North Dakota, South Dakota, Wisconsin NEWE Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont NY* New York* (NYCW, NYLI, and NYUP markets) RFCE Delaware, Maryland, New Jersey, Pennsylvania RFCM Michigan RFCW Illinois, Indiana, Ohio, West Virginia, Pennsylvania SRDA Arkansas, Louisiana, Mississippi SRGW Illinois, Missouri SRSE Alabama, Georgia, Mississippi, Florida SRCE Alabama, Kentucky, Mississippi, Tennessee SRVC North Carolina, South Carolina, Virginia SPNO Missouri, Kansas SPSO Arkansas, Louisiana, New Mexico, Oklahoma, Texas AZNM Arizona, California, Nevada, New Mexico CAMX California NWPP Idaho, Montana, Nevada, Oregon, Utah, Washington, Wyoming RMPA Colorado, Nebraska, South Dakota, Wyoming Source: Figure 4: Constructed by author based on EIA (2011a). Note: New York figures reflect the New York City Westchester electricity market. The 2015 Rate with CES estimates represent upper bounds that reflect several conservative assumptions: (1) no new generation capacity comes online by 2015; (2) regional emission intensities do not improve relative to their 2008 levels; and, (3) combustion turbine natural gas plants serve as marginal sources of production (and hence determine marginal cost pricing) in the ERCT, NEWE, NY*, RFCE, and CAMX regions. 16 Promoting Clean Energy in the American Power Sector

19 Revenue Raised through the Federal Clean Energy Credit. The federal clean energy credit ramps up from an initial price of $15 in 2015, and, given recent analyses of the potential for increasing the share of clean energy in U.S. power generation, it will very likely bind in (at least) the early years of the program. This provides a revenue source that could support investment in advanced energy R&D to promote the development of lower-cost technologies that could accelerate the eventual commercialization and deployment of clean energy. Some revenue could also address the potential increase in federal outlays to the extent the NCES increases electricity prices. The balance of the revenue could finance deficit reduction or a reduction in existing tax rates (e.g. reduce the payroll tax rate). To illustrate the potential magnitude of the revenue generation through the clean energy credit, recent carbon pricing analyses by the EIA suggest that the emission intensity of the U.S. power sector could fall to about 0.50 tco 2 /MWh in 2015 at a $15 credit price (EIA 2010b). Thus, the 0.40 tco 2 /MWh performance goal is a stretch for the economy in 2015, and sale of federal clean energy credits would net the government about $6.5 billion in revenues in 2015 (Figure 5). Other analyses that suggest that more clean energy could be deployed at this price (e.g., if wind or natural gas prices, or both, are lower) would deliver lower revenues. Given estimated improvement in emission intensity over the first decade of the NCES at the binding federal clean energy credit price, revenues are expected to increase to about $10 billion in 2020 and reach about $18 billion in These revenues would easily cover the proposed funding levels of $2 billion for a Clean Energy Fund in 2015 that ramps up to $5 billion in Impacts on manufacturing. The modest increase in electricity prices should have a limited impact on major industrial sector consumers of electricity. To illustrate these modest impacts, I have drawn from previous work that estimated the relationship between electricity prices and production and trade for more than four hundred U.S. manufacturing industries over twenty years (Aldy and Pizer 2009). This analysis allows for the effect of an increase in electricity price on an industry s production and on an industry s consumption (measured as production plus net imports) to vary with the energy intensity of manufacturing. The empirical model illustrates that the production from energy-lean manufacturing representing up to about the first 80 percent of the manufacturing sector when measured by energy intensity of output does not meaningfully change as electricity prices increase. More energy-intensive manufacturing, however, has historically experienced a decline in production and an increase in net imports when electricity prices have increased. FIGURE 5 Revenues Generated Through the Sale of Federal Clean Energy Credits, ($ billions) Source: Constructed by author based on estimated emission intensity and electricity generation under various scenarios presented in EIA (2010b). The Hamilton Project Brookings 17

20 I have employed our empirical model to simulate the impacts of a $15 federal clean energy credit price on select energyintensive manufacturing industries and the manufacturing sector as a whole (Table 5). Since the implicit generation subsidy mutes the price impact of a clean energy standard, the impact of this policy on industrial sector electricity prices is lower than it would be under a $15 per ton CO 2 cap-and-trade allowance price (the basis for the initial simulation presented in Aldy and Pizer 2009). The average national increase in the retail price of electricity in 2015 under the NCES would be about 0.25 /kwh, representing about a 4.1 percent increase in industrial sector electricity rates. 7 The manufacturing sector as a whole would experience about a 0.7 percent decline in production, with about half of this (0.4 percent) made up by an increase in net imports. For energy-intensive manufacturing, the decline in production is greater than the average for manufacturing ranging between 0.9 and 1.9 percent for the six major classes of energy-intensive manufacturing listed in the table but is relatively small within the context of historic annual swings in manufacturing output. Across the energyintensive manufacturing industries, the competitiveness effect the increase in net imports ranges from about 0.4 to 0.6 percent. This net import effect is also relatively small within the context of historic annual swings in the U.S. trade position. The start-up of the NCES in 2015 with a federal clean energy credit price set at $15 would appear to have only very modest effects on energy-intensive manufacturing. TABLE 5 Estimated Impacts of a $15 Clean Energy Credit Price on Energy-Intensive Manufacturing in 2015 Industry Production Consumption Competitiveness Industrial Chemicals 1.5% 1.0% 0.5% Paper 1.8% 1.3% 0.6% Iron & Steel 1.5% 1.1% 0.5% Aluminum 1.1% 0.7% 0.4% Cement 0.9% 0.5% 0.4% Bulk Glass 1.9% 1.5% 0.4% Manufacturing average 0.7% 0.3% 0.4% Source: Constructed by author based on Aldy and Pizer (2009). Note: Simulation results based on the assumption that a $15 credit price raises electricity prices in the industrial sector by 4 percent on average nationally. Competitiveness = Production Consumption. 18 Promoting Clean Energy in the American Power Sector

21 CLEAN ENERGY DEPLOYMENT AND EMISSION IMPACTS If U.S. clean energy generation delivers an emission intensity consistent with the proposed 2035 standard, then power sector emissions are likely to be about 60 percent below their 2005 levels (Figure 6). The 2020 performance goal for that year of 0.35 tco2/mwh would result in power sector emissions about 34 percent below 2005 levels. An illustration of the potential impact of the federal clean energy credit on emissions drawn by the author from past carbon price modeling analyses by the EIA (2010b) suggests that the $21 credit price in 2020 would result in power sector emissions of 2.04 billion metric tons of CO 2 15 percent below 2005 levels (and equal to about 13 percent below the no new policy reference case). FIGURE 6 U.S. Power Sector Carbon Pollution and Estimated Through 2035 with National Clean Energy Standard Performance Goals Million Metric Tons of CO 2 Source: Constructed by author based on EIA (2010b). Note: The estimates reflect an assumption that the performance goals are met without accessing the federal clean energy credits and a conservative assumption that generation levels do not fall relative to the expectation under the status quo. The Hamilton Project Brookings 19